US11462102B2 - Storage medium, estimation device, and estimation method - Google Patents

Storage medium, estimation device, and estimation method Download PDF

Info

Publication number
US11462102B2
US11462102B2 US16/827,786 US202016827786A US11462102B2 US 11462102 B2 US11462102 B2 US 11462102B2 US 202016827786 A US202016827786 A US 202016827786A US 11462102 B2 US11462102 B2 US 11462102B2
Authority
US
United States
Prior art keywords
information
indefiniteness
measurement data
points
traffic volume
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Active, expires
Application number
US16/827,786
Other languages
English (en)
Other versions
US20200226923A1 (en
Inventor
Toshio Endoh
Kazuhiro Matsumoto
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Fujitsu Ltd
Original Assignee
Fujitsu Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Fujitsu Ltd filed Critical Fujitsu Ltd
Assigned to FUJITSU LIMITED reassignment FUJITSU LIMITED ASSIGNMENT OF ASSIGNORS INTEREST (SEE DOCUMENT FOR DETAILS). Assignors: ENDOH, TOSHIO, MATSUMOTO, KAZUHIRO
Publication of US20200226923A1 publication Critical patent/US20200226923A1/en
Application granted granted Critical
Publication of US11462102B2 publication Critical patent/US11462102B2/en
Active legal-status Critical Current
Adjusted expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G08—SIGNALLING
    • G08G—TRAFFIC CONTROL SYSTEMS
    • G08G1/00—Traffic control systems for road vehicles
    • G08G1/01—Detecting movement of traffic to be counted or controlled
    • G08G1/0104—Measuring and analyzing of parameters relative to traffic conditions
    • G08G1/0137—Measuring and analyzing of parameters relative to traffic conditions for specific applications
    • G08G1/0145—Measuring and analyzing of parameters relative to traffic conditions for specific applications for active traffic flow control
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
    • G06Q10/047—Optimisation of routes or paths, e.g. travelling salesman problem
    • G—PHYSICS
    • G08—SIGNALLING
    • G08G—TRAFFIC CONTROL SYSTEMS
    • G08G1/00—Traffic control systems for road vehicles
    • G08G1/01—Detecting movement of traffic to be counted or controlled
    • G08G1/0104—Measuring and analyzing of parameters relative to traffic conditions
    • G08G1/0108—Measuring and analyzing of parameters relative to traffic conditions based on the source of data
    • G08G1/0116—Measuring and analyzing of parameters relative to traffic conditions based on the source of data from roadside infrastructure, e.g. beacons
    • G—PHYSICS
    • G08—SIGNALLING
    • G08G—TRAFFIC CONTROL SYSTEMS
    • G08G1/00—Traffic control systems for road vehicles
    • G08G1/01—Detecting movement of traffic to be counted or controlled
    • G08G1/0104—Measuring and analyzing of parameters relative to traffic conditions
    • G08G1/0125—Traffic data processing
    • G08G1/0129—Traffic data processing for creating historical data or processing based on historical data
    • G—PHYSICS
    • G08—SIGNALLING
    • G08G—TRAFFIC CONTROL SYSTEMS
    • G08G1/00—Traffic control systems for road vehicles
    • G08G1/123—Traffic control systems for road vehicles indicating the position of vehicles, e.g. scheduled vehicles; Managing passenger vehicles circulating according to a fixed timetable, e.g. buses, trains, trams
    • G08G1/127—Traffic control systems for road vehicles indicating the position of vehicles, e.g. scheduled vehicles; Managing passenger vehicles circulating according to a fixed timetable, e.g. buses, trains, trams to a central station ; Indicators in a central station

Definitions

  • the present invention relates to a storage medium, an estimation device, and an estimation method.
  • OD information refers to a traffic volume from one representative point (point) to another representative point (point).
  • Examples of an OD information estimation method include a method for estimating OD information by using survey (questionnaire) results for users who use a target traffic facility in a traffic system, traffic behavior histories recorded on IC cards of the users, and the like.
  • OD information estimation method a method for estimating OD information by combining the traffic behavior histories acquired from the users and congestion information.
  • International Publication Pamphlet No. WO 2014/203391 or the like is disclosed as related art.
  • a non-transitory computer-readable storage medium storing a program that causes a computer to execute a process, the process comprising: calculating parameters of an expression that describes a model that is associated with a traffic system and estimates a traffic volume between a plurality of points, by using a plurality of measurement data measured in the traffic system that connects the plurality of points to each other, by using a method for solving an optimization problem; generating, when an indefinite solution is obtained by the calculating, indefiniteness information related to a range of the indefinite solution in the method for solving the optimization problem; and determining addition of the plurality of measurement data or output of the indefinite solution, depending on the indefiniteness information.
  • FIG. 1 is a functional block diagram illustrating a configuration of an estimation device according to an embodiment.
  • FIG. 2 is a diagram illustrating an example of a target railroad route map.
  • FIG. 3 is a diagram illustrating an example of a target train diagram.
  • FIG. 4 is a diagram illustrating an example of target measurement data information.
  • FIG. 5 is a diagram illustrating a specific example of the measurement data information from 0:00 to 1:00.
  • FIG. 6 is a diagram illustrating examples of expressions used in an expression conversion unit.
  • FIG. 7 is a diagram illustrating a specific example of an optimization unit.
  • FIG. 8 is a diagram illustrating an existence range of solutions.
  • FIG. 9A is a diagram (1) illustrating influence of indefiniteness of OD information.
  • FIG. 9B is a diagram (2) illustrating the influence of the indefiniteness of the OD information.
  • FIG. 9C is a diagram (3) illustrating the influence of the indefiniteness of the OD information.
  • FIG. 9D is a diagram (4) illustrating the influence of the indefiniteness of the OD information.
  • FIG. 10 is a diagram illustrating an example of a flowchart of estimation processing according to the embodiment.
  • FIG. 11 is a diagram illustrating an example of a flowchart of processing of calculating a measure of indefiniteness of a solution according to the embodiment.
  • FIG. 12 is a diagram illustrating an example of a computer that executes an estimation program.
  • OD information is affected by a traffic use environment such as weather, day of the week, and presence of an event when traffic in a traffic system is used, so the OD information is desirably estimated by using a small amount of data that is least affected.
  • measures are required such as increasing the data used for estimation.
  • it is difficult to estimate appropriate OD information because OD information estimation is not performed in consideration of these contradictory constraints.
  • FIG. 1 is a diagram illustrating an example of a functional configuration of an estimation device according to the embodiment.
  • an estimation device 1 estimates OD information of a traffic system by using available data.
  • OD information refers to a traffic volume from one representative point (point) to another representative point (point).
  • a railroad traffic system will be described as an example; however, the present invention is not limited to the example, and is also applicable to general traffic systems such as buses and highways.
  • the estimation device 1 includes a control unit 10 and a storage unit 20 .
  • the control unit 10 corresponds to an electronic circuit such as a central processing unit (CPU). Then, the control unit 10 includes an internal memory for storing programs defining various processing procedures and control data, and executes various types of processing using the programs and the control data.
  • the control unit 10 includes an input unit 11 , an expression conversion unit 12 , an optimization unit 13 , an indefiniteness determination unit 14 , an expression correction unit 15 , and an output unit 16 .
  • the expression conversion unit 12 and the optimization unit 13 are examples of a calculation unit.
  • the indefiniteness determination unit 14 is an example of a generation unit and a determination unit.
  • the storage unit 20 is, for example, a semiconductor memory device such as a random access memory (RAM) and a flash memory, or a storage device such as a hard disk or an optical disk.
  • the storage unit 20 includes, for example, measurement data information 21 .
  • the measurement data information 21 is information including a plurality of data measured in a traffic system.
  • Examples of the measurement data information 21 include the number of entering people and the number of exiting people at each station, and the number of passengers on a train between adjacent stations.
  • the number of entering people and the number of exiting people at each station can be measured by a camera or a weight scale installed at each station.
  • the number of passengers on the train between adjacent stations can be measured by a camera or a weight scale installed in each vehicle of the train. It is expected that the number of entering people and the number of exiting people at each station and the number of passengers on the train between adjacent stations are available measurement data for any traffic system.
  • the station is an example of a representative point or a point.
  • the number of entering people is an example of a generated traffic volume.
  • the number of exiting people is an example of a concentrated traffic volume.
  • the number of passengers is an example of an allocated traffic volume.
  • the input unit 11 inputs the measurement data information 21 .
  • a specific example will be described later of the measurement data information 21 .
  • the expression conversion unit 12 converts each data of the measurement data information 21 input by the input unit 11 into an expression that describes a model that is associated with a traffic system and estimates a traffic volume between a plurality of stations.
  • the expression uses OD information as parameters (unknown numbers). An example will be described later of the expression used in the expression conversion unit 12 .
  • the optimization unit 13 calculates parameters of the expression converted by the expression conversion unit 12 by using a method for solving an optimization problem. Furthermore, the optimization unit 13 calculates parameters of an expression corrected by the expression correction unit 15 described later by using the method for solving the optimization problem.
  • the method for solving the optimization problem is, for example, linear programming, but is not limited thereto. In the embodiment, the method for solving the optimization problem will be described as linear programming. Furthermore, a specific example will be described later of the optimization unit 13 .
  • the indefiniteness determination unit 14 determines indefiniteness of a solution optimized by the optimization unit 13 .
  • the solution here indicates OD information.
  • the indefiniteness determination unit 14 calculates a measure of the indefiniteness of the solution related to an existence range of the solutions.
  • the indefiniteness determination unit 14 determines whether or not the measure of the indefiniteness of the solution is greater than or equal to a predetermined threshold value. If the measure of the indefiniteness of the solution is greater than or equal to the threshold value, the indefiniteness determination unit 14 determines that the indefiniteness of the solution has an influence, and transitions to the expression correction unit 15 .
  • the indefiniteness determination unit 14 determines that the indefiniteness of the solution does not have an influence, and outputs the solution to the output unit 16 . Descriptions will be given later of the measure of the indefiniteness, the existence range of the solutions, and the influence of the indefiniteness of the solution.
  • the expression correction unit 15 corrects the expression by increasing the measurement data information 21 .
  • the output unit 16 outputs OD information obtained from the solution. To prevent processing from being repeated indefinitely, in a case where the indefiniteness of the solution does not become less than the threshold value after a predetermined number of repetitions, the output unit 16 may output that fact.
  • a railroad route map and a railroad diagram taken up in the embodiment will be described with reference to FIGS. 2 and 3 .
  • FIG. 2 is a diagram illustrating an example of a target railroad route map. As illustrated in FIG. 2 , a target railroad route is straight, and includes three stations, A station, B station, and C station.
  • the railroad route map illustrated in FIG. 2 is an example for convenience, and the railroad route map is not limited to the example.
  • FIG. 3 is a diagram illustrating an example of a target train diagram. As illustrated in FIG. 3 , in the target train diagram, trains operate in both directions at 0:00, 1:00, and 2:00 p.m. regularly.
  • the train diagram illustrated in FIG. 3 is an example for convenience, and the train diagram is not limited to the example.
  • FIG. 4 is a diagram illustrating an example of target measurement data information.
  • the number of entering people and the number of exiting people at each station and the number of passengers on a train between adjacent stations are measured by using, for example, a camera, a weight scale, or the like.
  • the number of entering people at A station is defined as G A .
  • the number of exiting people at A station is defined as A A .
  • the number of entering people at B station is defined as G B .
  • the number of exiting people at B station is defined as A B .
  • the number of entering people at C station is defined as G C .
  • the number of exiting people at C station is defined as A C .
  • the number of passengers on a train from A station to B station (between A and B) is defined as P AB .
  • the number of passengers on a train from B station to A station (between B and A) is defined as P BA .
  • the number of passengers on a train from B station to C station (between B and C) is defined as P BC .
  • the number of passengers on a train from C station to B station (between C and B) is defined as P CB . It is assumed that the train operation time is sufficiently short and movement of users does not occur during train operation. Furthermore, it is also assumed that the users get on the earliest train after entering the station.
  • FIG. 5 is a diagram illustrating a specific example of the measurement data information from 0:00 to 1:00.
  • the G A indicating the number of entering people at A station is set to 50.
  • the G B indicating the number of entering people at B station is set to 70.
  • the G C indicating the number of entering people at C station is set to 60.
  • the A A indicating the number of exiting people at A station is set to 70.
  • the A B indicating the number of exiting people at B station is set to 40.
  • the A C indicating the number of exiting people at C station is set to 70.
  • the P AB indicating the number of passengers on the train between A and B is set to 50.
  • the P BC indicating the number of passengers on the train between B and C is set to 70.
  • the P BA indicating the number of passengers on the train between B and A is set to 70.
  • the number of passengers P CB indicating the number of passengers on the train between C and B is set to 60.
  • the measurement data information 21 from 0:00 to 1:00 may be indicated by a reference sign D 01 hereinafter.
  • FIG. 6 is a diagram illustrating the examples of the expressions used in the expression conversion unit.
  • T AB illustrated in FIG. 6 is defined as OD information from 0:00 to 1:00 from A station to B station.
  • T AC illustrated in FIG. 6 is defined as OD information from 0:00 to 1:00 from A station to C station.
  • T BA illustrated in FIG. 6 is defined as OD information from 0:00 to 1:00 from B station to A station.
  • T BC illustrated in FIG. 6 is defined as OD information from 0:00 to 1:00 from B station to C station.
  • T CA illustrated in FIG. 6 is defined as OD information from 0:00 to 1:00 from C station to A station.
  • T CB illustrated in FIG. 6 is defined as OD information from 0:00 to 1:00 from C station to B station. As illustrated in FIG.
  • the expression conversion unit 12 converts the measurement data information 21 into an expression having the OD information as an unknown number.
  • the conversion method is as follows. (1) Conversion is performed into a constraint expression in which the number of entering people at a station is equal to the sum of OD information with the station as the departure station. (2) Conversion is performed into a constraint expression in which the number of exiting people at a station is equal to the sum of OD information with the station as the arrival station. (3) The number of passengers is converted into an evaluation expression including an absolute value of a difference between the number of passengers predicted from the OD information and the number of passengers as measurement data. (4) The sum of evaluation expressions is defined as an evaluation function.
  • the numbers of entering people at the stations are converted into the following constraint expressions. Conversion is performed into a constraint expression (1.1) in which the G A indicating the number of entering people at A station is equal to the sum of the OD information T AB between A and B and the OD information T AC between A and C. Conversion is performed into a constraint expression (1.2) in which the G B indicating the number of entering people at B station is equal to the sum of the OD information T BA between B and A and the OD information T BC between B and C. Conversion is performed into a constraint expression (1.3) in which the G C indicating the number of entering people at C station is equal to the sum of the OD information T CA between C and A and the OD information T CB between C and B.
  • the numbers of exiting people at the stations are converted into the following constraint expressions. Conversion is performed into a constraint expression (2.1) in which the A A indicating the number of exiting people at A station is equal to the sum of the OD information T BA between B and A and the OD information T CA between C and A with A station as the arrival station. Conversion is performed into a constraint expression (2.2) in which the A B indicating the number of exiting people at B station is equal to the sum of the OD information T AB between A and B and the OD information T CB between C and B with B station as the arrival station.
  • Conversion is performed into a constraint expression (2.3) in which the A C indicating the number of exiting people at C station is equal to the sum of the OD information T AC between A and C and the OD information T BC between B and C with C station as the arrival station.
  • the numbers of passengers are converted into the following evaluation expressions.
  • the number of passengers is converted into an evaluation expression (3.1) including an absolute value of a difference between the P AB indicating the number of passengers on the train between A and B as measurement data and the numbers of passengers T AB and T AC predicted from the OD information.
  • the number of passengers is converted into an evaluation expression (3.2) including an absolute value of a difference between the P BC indicating the number of passengers on the train between B and C as measurement data and the numbers of passengers T AC and T BC predicted from the OD information.
  • the number of passengers is converted into an evaluation expression (3.3) including an absolute value of a difference between the P BA indicating the number of passengers on the train between B and A as measurement data and the numbers of passengers T BA and T CA predicted from the OD information.
  • the number of passengers is converted into an evaluation expression (3.4) including an absolute value of a difference between the P CB indicating the number of passengers on the train between C and B as measurement data and the numbers of passengers T CA and T CB predicted from the OD information.
  • an evaluation function is generated as follows. The sum of the individual evaluation expressions (3.1), (3.2), (3.3), and (3.4) is defined as an evaluation function (4). Furthermore, the numbers of passengers T AB , T AC , T BA , T BC , T CA , and T CB predicted are all greater than or equal to 0.
  • Each constraint expression is a linear expression with a non-negative condition.
  • the evaluation function (4) is a piecewise linear function convex downward.
  • Each constraint expression and the evaluation function (4) are expressed as a minimization problem. This becomes a linear programming problem.
  • each constraint expression and the evaluation function (4) are expressed as a standard linear programming problem expressed by an expression (5).
  • the symbol A is a matrix represented by each constraint expression and the evaluation function (4).
  • the symbol x is an unknown column vector, and is a solution.
  • the symbol b is a column vector represented by the measurement data information 21 .
  • “0 v ” is a column vector in which all elements are 0.
  • “ ⁇ ” is comparison for each element of the vector.
  • the symbol T represents transposition of a matrix.
  • the symbol c is a column vector.
  • the x, b, c, and A are expressed by expressions (6) to (9), respectively.
  • the F AB in the expression (6) corresponds to the evaluation expression (3.1).
  • the F BC in the expression (6) corresponds to the evaluation expression (3.2).
  • the F BA in the expression (6) corresponds to the evaluation expression (3.3).
  • the F CB in the expression (6) corresponds to the evaluation expression (3.4).
  • the optimization unit 13 solves the linear programming problem expressed by the expression (5), and calculates the solution x that minimizes the evaluation function (4) including the evaluation expressions (3.1) to (3.4). In other words, the optimization unit 13 solves the given linear programming problem and calculates a value of the OD information that minimizes the evaluation function.
  • FIG. 7 is a diagram illustrating a specific example of the optimization unit.
  • x (0,50,50,20,20,40,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0) T
  • the value of the evaluation function (4) is 0. From the first element of x, the OD information T AB between A and B is 0. From the second element of x, the OD information T AC between A and C is 50. From the third element of x, the OD information T BA between B and A is 50. From the fourth element of x, the OD information T BC between B and C is 20. From the fifth element of x, the OD information T CA between C and A is 20. From the sixth element of x, the OD information T CB between C and B is 40.
  • the OD information from 0:00 to 1:00 obtained as a solution is illustrated in the upper diagram of FIG. 7 .
  • the OD information T AB from 0:00 to 1:00 from A station to B station is 0.
  • the OD information T AC from 0:00 to 1:00 from A station to C station is 50.
  • the OD information T BA from 0:00 to 1:00 from B station to A station is 50.
  • the OD information T BC from 0:00 to 1:00 from B station to C station is 20.
  • the OD information T CA from 0:00 to 1:00 from C station to A station is 20.
  • the OD information T CB from 0:00 to 1:00 from C station to B station is 40.
  • x (40,10,10,60,60,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0) T
  • the OD information from 0:00 to 1:00 obtained as a solution is illustrated in the lower diagram of FIG. 7 .
  • the solution optimized by the given linear programming problem is not unique.
  • the solution is an indefinite solution.
  • the indefiniteness determination unit 14 calculates the measure of the indefiniteness of the solution related to the existence range of the solutions.
  • a method will be described for determining the uniqueness of the solution of the linear programming problem and calculating the measure of the indefiniteness of the solution.
  • the given linear programming problem is expressed in a standard form as illustrated in the expression (5).
  • the solution obtained by the optimization unit 13 is defined as x 0 .
  • a set of subscripts in which values of the elements are 0 among the elements of x 0 is defined as K, and a set of subscripts in which values of the elements are positive is defined as J.
  • a matrix with c T added to the bottom row of A is defined as A′. The following theorem is used for determination of the uniqueness of the solution of the linear programming problem and calculation of the degree of the indefiniteness of the solution.
  • x 0 is unique is equivalent to the following condition.
  • Column vectors of A′ J are linearly independent of each other, and an absolute value norm (sum of absolute values of elements)
  • of x that maximizes 1 v T x K under constraint conditions of A′x 0 v , and x ⁇ x 0 , is 0.
  • A′ J is a submatrix obtained by extracting only a column vector of an element J from A′.
  • the symbol 0 v is a column vector in which all elements are 0.
  • the symbol x K is a column vector obtained by extracting only an element K from x.
  • the symbol 1 v is a column vector in which all elements are 1.
  • x J is a column vector obtained by extracting only an element J from x.
  • x K is not 0 v or that the column vectors of A′ J are linearly dependent on each other.
  • x K is not 0 v
  • x that gives the maximum value of 1 v T x K is not 0 v .
  • FIG. 8 is a diagram illustrating the existence range of the solutions. As illustrated in FIG. 8 , because of convexity of the solution, all points on a line segment connecting two points x 0 and x 0 +x are also solutions.
  • the length of this line segment can be obtained by, for example, the absolute value norm
  • an average value divided by the number of unknown numbers of the OD information is more preferable. Therefore, in the embodiment, as the measure of the indefiniteness of the solution of the given linear programming problem, a value is used obtained by dividing the absolute value norm
  • the fact that the measure of the indefiniteness is 0 is equivalent to that the solution is unique. Since the degree of the indefiniteness of the solution can be quantitatively evaluated, it is possible to distinguish between a case where the indefiniteness of the solution is small and does not have an influence, and a case where the indefiniteness of the solution is large and has an influence.
  • the indefiniteness determination unit 14 corrects the solution so that the column vectors of A′ J become linearly independent of each other and then calculates the measure.
  • the column vectors are caused to be linearly independent of each other by increasing elements taking a value 0 in the solution to reduce the number of column vectors of A′ J .
  • maximization of 1 v T x K is a linear programming problem, it can be solved by using a method of a conventional technique.
  • the indefiniteness determination unit 14 applies the method described above to the problem taken up in FIGS. 2 to 7 , to calculate the measure of the indefiniteness of the solution.
  • the number of unknown numbers of the OD information is six.
  • x 0 is expressed as follows (upper diagram in FIG. 7 ).
  • x 0 (0,50,50,20,20,40,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0) T
  • the indefiniteness determination unit 14 obtains, from the elements of x 0 , a set of subscripts J in which values of the elements are positive, and a set of subscripts K in which values of the elements are 0, as follows. Subscripts are counted from one.
  • the indefiniteness determination unit 14 obtains A′ J as in an expression (10).
  • Elements of x are represented by x 1 , x 2 , . . . , x 18 .
  • x x 1 (1, ⁇ 1, ⁇ 1,1,1, ⁇ 1,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0) T , x 1 ⁇ 0, x 1 ⁇ 40
  • the evaluation function 1 v T x K is equal to x 1 .
  • the indefiniteness determination unit 14 obtains the maximum value of this linear programming problem as 40, and obtains the absolute value norm
  • the value 40 is equal to a difference between elements of the OD information in the upper diagram of FIG. 7 and the lower diagram of FIG. 7 .
  • the indefiniteness determination unit 14 determines whether or not the measure of the indefiniteness is greater than or equal to a predetermined threshold value.
  • the threshold value is 1. Then, since the measure of the indefiniteness exceeds the threshold value, the indefiniteness determination unit 14 determines that the OD information cannot be determined with the given data. In other words, the indefiniteness determination unit 14 determines that the indefiniteness of the OD information has an influence.
  • the influence will be described of the indefiniteness of the OD information. Even if the OD information is not uniquely determined, the OD information satisfies the constraint expressions obtained from the measurement data information 21 and minimizes the evaluation function, so if a traffic use environment does not change, the OD information can be used for congestion prediction and traffic jam prediction of the traffic system. However, for example, in a case where an abnormality occurs such as an accident, an unrealistic prediction result may be obtained in which accuracy is degraded due to the indefiniteness of the OD information. Thus, with the railroad route map taken up in FIG. 2 as an example, taking up a case where the train is delayed, it will be described that the indefiniteness of the OD information has an influence, in other words, causes degradation of prediction accuracy.
  • FIGS. 9A to 9D are diagrams illustrating the influence of the indefiniteness of the OD information.
  • correct answer OD information from 0:00 to 1:00 and correct answer OD information from 1:00 to 2:00 are obtained.
  • Each piece of correct answer OD information can be obtained by, for example, a survey (questionnaire) for users who use the target railroad.
  • the upper left diagram in FIG. 9A illustrates the correct answer OD information from 0:00 to 1:00.
  • the correct answer OD information from 0:00 to 1:00 is referred to as a correct answer OD 01 .
  • the upper right diagram in FIG. 9A illustrates the correct answer OD information from 1:00 to 2:00.
  • the correct answer OD information from 1:00 to 2:00 is referred to as a correct answer OD 12 .
  • the measurement data information 21 is calculated indicating the number of entering and exiting people at each station from 1:00 to 2:00 and the number of passengers on a train at 2:00.
  • the middle diagram in FIG. 9A illustrates the number of entering and exiting people at each station from 1:00 to 2:00 and the number of passengers on the train at 2:00.
  • the number of entering people G A at A station is 70.
  • the number of exiting people A A at A station is 40.
  • the number of entering people G B at B station is 50.
  • the number of exiting people A B at B station is 70.
  • the number of entering people G C at C station is 60.
  • the number of exiting people A C at C station is 70.
  • the number of passengers P AB on the train from A station to B station is 70.
  • the number of passengers P BA on the train from B station to A station is 40.
  • the number of passengers P BC on the train from B station to C station is 70.
  • the number of passengers P CB on the train from C station to B station is 60.
  • the measurement data information 21 indicating the number of entering people at each station from 1:00 to 2:00 and the number of passengers on the train at 2:00 may be indicated by a reference sign D 12 .
  • the number of entering people at each station from 0:00 to 1:00 and the number of passengers on a train at 1:00 are as indicated by the reference sign D 01 in FIG. 5 .
  • the measurement data information 21 illustrated in the middle of FIG. 9A is subjected to processing by the expression conversion unit 12 and processing by the optimization unit 13 , whereby the OD information is calculated.
  • the lower diagram in FIG. 9A illustrates the OD information from 1:00 to 2:00 obtained as a solution.
  • the OD information T AB from 1:00 to 2:00 from A station to B station is 50.
  • the OD information T AC from 1:00 to 2:00 from A station to C station is 20.
  • the OD information T BA from 1:00 to 2:00 from B station to A station is 0.
  • the OD information T BC from 1:00 to 2:00 from B station to C station is 50.
  • the OD information T CA from 1:00 to 2:00 from C station to A station is 40.
  • the OD information T CB from 1:00 to 2:00 from C station to B station is 20.
  • the OD information from 1:00 to 2:00 obtained as the solution is referred to as a solution OD 12 .
  • the OD information (a solution OD 01 ) from 0:00 to 1:00 obtained as a solution is as illustrated in the upper diagram of FIG. 7 .
  • FIG. 9B The upper diagram of FIG. 96 illustrates a delay pattern x that is a case where the train is delayed by 30 minutes at A station and departs from A station at 1:30.
  • the lower diagram of FIG. 9B illustrates a delay pattern y that is a case where the train is delayed by 30 minutes at B station and departs from B station at 1:30.
  • the user behaves as follows. The user uses the railroad in accordance with the correct answer OD information. The user arrives at the station uniformly at each time zone. Even if the train is delayed, the user does not take any other action and waits for the delayed train.
  • correct answers of the number of entering and exiting people at each station and the number of passengers on each train are calculated by using the correct answer OD 01 and the correct answer OD 12 .
  • the number of entering and exiting people at each station from 0:00 to 1:00 and the number of passengers on the train at 1:00 and 1:30 are calculated (Dx 01 ).
  • the number of entering and exiting people at each station from 1:00 to 2:00 and the number of passengers on the train at 2:00 are calculated (Dx 12 ).
  • the number of entering and exiting people at each station from 0:00 to 1:00 and the number of passengers on the train at 1:00 and 1:30 are calculated (Dy 01 ).
  • the number of entering and exiting people at each station from 1:00 to 2:00 and the number of passengers on the train at 2:00 are calculated (Dy 12 ).
  • the number of entering and exiting people at each station and the number of passengers on each train are calculated by using the OD information obtained as the solutions (the solution OD 01 and the solution OD 12 ).
  • the number of entering and exiting people at each station from 0:00 to 1:00 and the number of passengers on the train at 1:00 and 1:30 are calculated (Dx 01 ′).
  • the number of entering and exiting people at each station from 1:00 to 2:00 and the number of passengers on the train at 2:00 are calculated (Dx 12 ′).
  • the number of entering and exiting people at each station from 0:00 to 1:00 and the number of passengers on the train at 1:00 and 1:30 are calculated (Dy 01 ′).
  • the number of entering and exiting people at each station from 1:00 to 2:00 and the number of passengers on the train at 2:00 are calculated (Dy 12 ′).
  • the correct answer for the delay pattern illustrated in FIG. 9C is compared with the estimated value for the delay pattern illustrated in FIG. 9D .
  • the correct answer Dx 01 for the delay pattern x and the estimated value Dx 01 ′ for the delay pattern x are compared with each other. Then, the number of exiting people A B at B station in the correct answer Dx 01 for the delay pattern x is 20, whereas the number of exiting people A B at B station in the estimated value Dx 01 ′ for the delay pattern x is 40. Thus, it can be seen that the number of exiting people A B at B station in the estimated value Dx 01 ′ for the delay pattern x is different from the correct answer Dx 01 for the delay pattern x.
  • the correct answer Dx 12 for the delay pattern x and the estimated value Dx 12 ′ for the delay pattern x are compared with each other. Then, the number of exiting people A B at B station in the correct answer Dx 12 for the delay pattern x is 90, whereas the number of exiting people A B in B station in the estimated value Dx 12 ′ for the delay pattern x is 70. Thus, it can be seen that the number of exiting people A B at B station in the estimated value Dx 12 ′ for the delay pattern x is different from the correct answer Dx 12 for the delay pattern x.
  • the correct answer Dy 01 for the delay pattern y and the estimated value Dy 01 ′ for the delay pattern y are compared with each other. Then, the number of passengers P BA on the train from B station to A station (between B and A) in the correct answer Dy 01 for the delay pattern y is 80, whereas the number of passengers P BA on the train between B and A in the estimated value Dy 01 ′ for the delay pattern y is 70. Thus, it can be seen that the number of passengers P BA on the train between B and A in the estimated value Dy 01 ′ for the delay pattern y is different from the correct answer Dy 01 for the delay pattern y.
  • the correct answer Dy 12 for the delay pattern y and the estimated value Dy 12 ′ for the delay pattern y are compared with each other. Then, the number of passengers P BA on the train between B and A in the correct answer Dy 12 for the delay pattern y is 30, whereas the number of passengers P BA on the train between B and A in the estimated value Dy 12 ′ for the delay pattern y is 40. Thus, it can be seen that the number of passengers P BA on the train between B and A in the estimated value Dy 12 ′ for the delay pattern y is different from the correct answer Dy 12 for the delay pattern y.
  • the indefiniteness of the OD information has an influence, in other words, causes degradation of the prediction accuracy.
  • the expression correction unit 15 corrects the expression by increasing the measurement data information 21 so that the indefiniteness of the OD information is reduced. Examples of a typical method include sharing of unknown numbers. The method defines common average OD information for a plurality of pieces of the measurement data information 21 . By using the measurement data information 21 measured in a similar traffic use environment, the expression correction unit 15 can suppress an error due to sharing the OD information low.
  • the measurement data information 21 currently used is the measurement data information 21 (D 01 ) from 0:00 to 1:00 illustrated in FIG. 5 . In other words, it is the number of entering and exiting people at each station from 0:00 to 1:00 and the number of passengers on the train at 1:00.
  • Examples of the measurement data information 21 to be increased include the measurement data information 21 (D 12 ) from 1:00 to 2:00 when there is no delay illustrated in the middle diagram of FIG. 9A , and the measurement data information 21 (Dx 01 , Dx 12 ) of the correct answer for the delay pattern x illustrated in FIG. 9C . In such an example, there are two types of the OD information of from 0:00 to 1:00 and from 1:00 to 2:00.
  • the expression correction unit 15 generates an expression in which unknown numbers are shared by combining the measurement data information 21 when there is no delay and the measurement data information 21 for the delay pattern x.
  • the number of unknown numbers is 12 including that in the case from 0:00 to 1:00 and that in the case from 1:00 to 2:00.
  • the optimization unit 13 solves the given linear programming problem for the expression corrected by the expression correction unit 15 , and calculates the value of the OD information that minimizes the evaluation function.
  • the calculated OD information from 0:00 to 1:00, matches the correct answer OD information (the correct answer OD 01 ) from 0:00 to 1:00 illustrated in FIG. 9A .
  • the calculated OD information from 1:00 to 2:00, matches the OD information (the solution OD 12 ) from 1:00 to 2:00 illustrated in FIG. 9A .
  • the indefiniteness determination unit 14 calculates the measure of the indefiniteness of the OD information that is the solution optimized by the optimization unit 13 .
  • the indefiniteness determination unit 14 calculates the existence range of the solutions.
  • the column vectors of A′ J are linearly independent of each other, and the absolute value norm
  • the indefiniteness determination unit 14 calculates a value 20 obtained by dividing the absolute value norm 240 by the number of unknown numbers 12 of the OD information, as the measure of OD information.
  • the indefiniteness determination unit 14 determines whether or not the measure of the indefiniteness is greater than or equal to a predetermined threshold value.
  • the threshold value is 1.
  • the indefiniteness determination unit 14 determines that the OD information cannot be determined with the given data. In other words, the indefiniteness determination unit 14 determines that there is an influence of the indefiniteness of the OD information.
  • the expression correction unit 15 corrects the expression by increasing the measurement data information 21 so that the indefiniteness of the OD information is reduced.
  • Examples of the measurement data information 21 to be increased include the measurement data information 21 (D 12 ) from 1:00 to 2:00 when there is no delay illustrated in the middle diagram of FIG. 9A , and the measurement data information 21 (Dx 01 , Dx 12 ) of the correct answer for the delay pattern x illustrated in FIG. 9C .
  • examples of the measurement data information 21 to be increased include the measurement data information 21 (Dy 01 , Dy 12 ) of the correct answer for the delay pattern y illustrated in FIG. 9C .
  • the optimization unit 13 solves the given linear programming problem for the expression corrected by the expression correction unit 15 , and calculates the value of the OD information that minimizes the evaluation function.
  • the calculated OD information from 0:00 to 1:00, matches the correct answer OD information (the correct answer OD 01 ) from 0:00 to 1:00 illustrated in FIG. 9A .
  • the calculated OD information from 1:00 to 2:00, matches the correct answer OD information (the correct answer OD 12 ) from 1:00 to 2:00 illustrated in FIG. 9A .
  • the indefiniteness determination unit 14 calculates the measure of the indefiniteness of the OD information that is the solution optimized by the optimization unit 13 .
  • the indefiniteness determination unit 14 calculates the existence range of the solutions.
  • the column vectors of A′ J are linearly independent of each other, and the absolute value norm
  • the indefiniteness determination unit 14 calculates a value 0 obtained by dividing the absolute value norm 0 by the number of unknown numbers of the OD information, as the measure of the OD information.
  • the indefiniteness determination unit 14 determines whether or not the measure of the indefiniteness is greater than or equal to a predetermined threshold value.
  • the threshold value is 1.
  • the indefiniteness determination unit 14 determines that the indefiniteness is within an allowable range and the OD information can be determined with the given data. In other words, the indefiniteness determination unit 14 determines that there is no influence of the indefiniteness of the OD information.
  • the output unit 16 outputs the calculated OD information.
  • FIG. 10 is a diagram illustrating an example of a flowchart of estimation processing according to the embodiment.
  • the input unit 11 reads the measurement data information 21 (step S 11 ).
  • the input unit 11 inputs the number of entering people and the number of exiting people at each station and the number of passengers on the train between adjacent stations for a predetermined period, as the measurement data information 21 .
  • the measurement data information 21 is information measured by using a camera, a weight scale, or the like.
  • the expression conversion unit 12 converts each data of the read measurement data information 21 into an expression that describes a model that is associated with a traffic system and estimates a traffic volume between a plurality of stations (step S 12 ).
  • the expression conversion unit 12 converts each data of the measurement data information 21 into an expression including the constraint expressions, the evaluation expressions, and the evaluation function illustrated in FIG. 6 .
  • Such an expression has the OD information as an unknown number.
  • the optimization unit 13 optimizes the linear programming problem including the expression converted by the expression conversion unit 12 (step S 13 ).
  • the optimization unit 13 expresses the converted constraint expressions, evaluation expressions, and evaluation function as a standard linear programming problem represented by the expression (5), and calculates x expressed by the expression (6).
  • the indefiniteness determination unit 14 calculates the measure of the indefiniteness of the solution optimized by the optimization unit 13 (step S 14 ). A flowchart will be described later of the processing of calculating the measure of the indefiniteness of the solution.
  • the indefiniteness determination unit 14 determines whether or not the measure of the indefiniteness of the solution is less than a threshold value (step S 15 ).
  • step S 15 determines that the OD information cannot be determined with the given data. In other words, the indefiniteness determination unit 14 determines that there is an influence of the indefiniteness of the OD information. Then, the expression correction unit 15 corrects the expression by increasing the measurement data information 21 so that the indefiniteness of the OD information is reduced (step S 16 ). Then, the expression correction unit 15 proceeds to step S 13 to cause the optimization unit 13 to perform optimization processing.
  • the indefiniteness determination unit 14 determines that the indefiniteness is within an allowable range and the OD information can be determined with the given data. In other words, the indefiniteness determination unit 14 determines that there is no influence of the indefiniteness of the OD information. Then, the output unit 16 outputs an optimized result as the OD information (step S 17 ). Then, the estimation processing ends.
  • FIG. 11 is a diagram illustrating an example of the flowchart of the processing of calculating the measure of the indefiniteness of the solution according to the embodiment.
  • the indefiniteness determination unit 14 determines whether or not the column vectors of A′ J are linearly independent of each other (step S 21 ).
  • A′ here refers to a matrix with c T added to the bottom row of A.
  • A′ J is a submatrix obtained by extracting only a column vector of an element J from A′. J indicates a set of subscripts in which values of the elements are positive among the elements of the solution x 0 .
  • step S 21 In a case where it is determined that the column vectors of A′ J are not linearly independent of each other (step S 21 ; No), the indefiniteness determination unit 14 corrects the solution so that the column vectors of A′ J become linearly independent of each other (step S 22 ). Then, the indefiniteness determination unit 14 proceeds to step S 21 .
  • the indefiniteness determination unit 14 calculates the absolute value norm of x that maximizes 1 v T x K (step S 23 ).
  • the indefiniteness determination unit 14 calculates a value obtained by dividing the calculated absolute value norm by the number of unknown numbers of the OD information, as the measure of the indefiniteness (step S 24 ). Then, the processing of calculating the measure of the indefiniteness of the solution ends.
  • the estimation device 1 calculates parameters of an expression that describes a model that is associated with a traffic system and estimates a traffic volume between a plurality of points, by using a plurality of measurement data measured in the traffic system that connects the plurality of points to each other, by using a method for solving an optimization problem.
  • the estimation device 1 generates indefiniteness information related to a range of the indefinite solution in the method for solving the optimization problem.
  • the estimation device 1 determines addition of the plurality of measurement data or output of the indefinite solution, depending on the indefiniteness information. According to such a configuration, the estimation device 1 can be used for estimation of the OD information by using the plurality of measurement data in the expression.
  • the estimation device is enabled to add the measurement data by using the indefiniteness information related to the range of the indefinite solution, and can estimate appropriate OD information by using the plurality of measurement data including the added measurement data in the expression. Furthermore, the estimation device 1 can estimate appropriate OD information by using available data in the traffic system. Furthermore, the estimation device 1 can estimate the OD information reflecting an evaluation of appropriateness of the model and indicating the traffic volume between the plurality of points, for the model associated with the traffic system in which it is difficult to obtain measurement data.
  • the estimation device 1 generates the indefiniteness information on the basis of the range of the indefinite solution in the method for solving the optimization problem and the number of the plurality of points. According to such a configuration, since the range of the indefinite solution increases with the number of the plurality of points, the estimation device 1 generates the indefiniteness information on the basis of the range of the indefinite solution and the number of the plurality of points, thereby being able to quantitatively obtain the measure of the range of the indefinite solution.
  • the measurement data are a generated traffic volume at each of the points, a concentrated traffic volume at each of the points, and an allocated traffic volume of a means of traffic.
  • the estimation device 1 can estimate appropriate OD information by using available measurement data in any traffic system.
  • the estimation device 1 converts the plurality of measurement data into a linear equality constraint in which a constraint of a total amount of OD information is equal to a given generated traffic volume.
  • the estimation device 1 converts the plurality of measurement data into a linear equality constraint in which the constraint of the total amount of the OD information is equal to a given concentrated traffic volume.
  • the estimation device 1 performs conversion into an evaluation function including a difference between an allocated traffic volume estimated from the OD information and a given allocated traffic volume.
  • the estimation device 1 calculates parameters related to the OD information of the converted expressions by using the method for solving the optimization problem.
  • the estimation device 1 can estimate the OD information by calculating the parameters related to the OD information by using the method for solving the optimization problem using the plurality of measurement data.
  • the method for solving the optimization problem is linear programming. According to such a configuration, the estimation device 1 can estimate the OD information at high speed by using linear programming.
  • the OD information is added up with reference to a departure point at a predetermined constant time interval (for example, 1 hour), but this is not a limitation.
  • the OD information may be added up with reference to the departure point at an arbitrary time interval, or may be added up with reference to an arrival point at a predetermined constant time interval or an arbitrary time interval.
  • the departure point reference is considered to be useful in a case where the user does not know the traffic environment change such as an accident.
  • the arrival point reference is considered to be useful in prediction of future traffic demand by which the user knows a change in the traffic environment.
  • the OD information may be estimated for each type of train such as a rapid train and a local train.
  • the expression conversion unit 12 performs conversion into the constraint expression regarding the number of entering and exiting people at the station, this is not a limitation. In a case where the number of entering and exiting people can be measured with a finer granularity than at the station, such as at a platform in the station, the expression conversion unit 12 may perform conversion into a constraint expression that takes the sum of the OD information depending on the granularity.
  • the expression conversion unit 12 assumes that the train operation time between stations is sufficiently short, but in a case where the train operation time between stations cannot be ignored, conversion may be performed into a constraint expression of the sum of the OD information considering the travel time on the train.
  • the expression conversion unit 12 may perform conversion into an evaluation expression on the assumption that passengers use the train giving the shortest time, for example. This corresponds to a case where a plurality of routes exists between two points, in a general traffic system.
  • the expression conversion unit 12 may use an evaluation expression that evaluates the number of passengers z predicted from the OD information with a function f(z) expressed as, for example, max (a ⁇ z, 0, z ⁇ b) instead of the absolute value of the difference.
  • the expression correction unit 15 corrects the expression by increasing the measurement data information 21 so that the indefiniteness of the OD information is reduced.
  • the expression correction unit 15 may correct the expression by adding a predetermined expression so that the indefiniteness of the OD information is reduced. For example, in a case where an average value ⁇ ij of the OD information from a station i to a station j is obtained from past statistics, the expression correction unit 15 corrects the expression by adding ⁇ ij
  • ⁇ is a predetermined positive number.
  • ⁇ ij indicates that the sum is obtained for all combinations of ij.
  • estimation device 1 can be implemented by mounting each function such as the control unit 10 and the storage unit 20 described above on an information processing device such as a known personal computer or workstation.
  • each component of the illustrated device is not necessarily physically configured as illustrated in the drawings.
  • a specific aspect of distribution and integration of the device is not limited to the illustrated one, and all or a part of the device can be functionally or physically distributed and integrated in an arbitrary unit depending on various loads, use states, and the like.
  • the input unit 11 and the output unit 16 may be integrated as one unit.
  • the indefiniteness determination unit 14 may be distributed into a first determination unit that determines whether or not the solution is an indefinite solution, a calculation unit that calculates a measure of indefiniteness in a case where the solution is an indefinite solution, and a second determination unit that determines whether or not the measure of the indefiniteness is greater than or equal to a threshold value.
  • the storage unit 20 may be connected as an external device of the estimation device 1 , or may be connected via a network.
  • FIG. 12 is a diagram illustrating an example of the computer that executes the estimation program.
  • a computer 200 includes a CPU 203 that executes various types of calculation processing, an input device 215 that accepts data input from a user, and a display control unit 207 that controls a display device 209 . Furthermore, the computer 200 includes a drive device 213 that reads a program and the like from a storage medium, and a communication control unit 217 that exchanges data with another computer via a network. Furthermore, the computer 200 includes a memory 201 that temporarily stores various types of information, and an HDD 205 .
  • the memory 201 , the CPU 203 , the HDD 205 , the display control unit 207 , the drive device 213 , the input device 215 , and the communication control unit 217 are connected to each other by a bus 219 .
  • the drive device 213 is a device for a removable disk 210 , for example.
  • the HDD 205 stores an estimation program 205 a and estimation related information 205 b.
  • the CPU 203 reads the estimation program 205 a , deploys the estimation program 205 a on the memory 201 , and executes the estimation program 205 a as a process. Such a process corresponds to each functional unit of the estimation device 1 .
  • the estimation related information 205 b corresponds to the measurement data information 21 , for example.
  • the removable disk 210 stores each piece of information such as the estimation program 205 a.
  • the estimation program 205 a does not necessarily have to be stored in the HDD 205 from the beginning.
  • the program is stored in a “portable physical medium” such as a flexible disk (FD), a CD-ROM, a DVD disk, a magneto-optical disk, or an IC card inserted in the computer 200 .
  • the computer 200 may read the estimation program 205 a from these media to execute the estimation program 205 a.

Landscapes

  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Engineering & Computer Science (AREA)
  • Business, Economics & Management (AREA)
  • Analytical Chemistry (AREA)
  • Chemical & Material Sciences (AREA)
  • Human Resources & Organizations (AREA)
  • Economics (AREA)
  • Strategic Management (AREA)
  • Radar, Positioning & Navigation (AREA)
  • Quality & Reliability (AREA)
  • Game Theory and Decision Science (AREA)
  • Remote Sensing (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Marketing (AREA)
  • Operations Research (AREA)
  • Development Economics (AREA)
  • Tourism & Hospitality (AREA)
  • General Business, Economics & Management (AREA)
  • Theoretical Computer Science (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)
  • Train Traffic Observation, Control, And Security (AREA)
  • Traffic Control Systems (AREA)
US16/827,786 2017-09-29 2020-03-24 Storage medium, estimation device, and estimation method Active 2038-01-06 US11462102B2 (en)

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/JP2017/035591 WO2019064526A1 (ja) 2017-09-29 2017-09-29 推定プログラム、推定装置および推定方法

Related Parent Applications (1)

Application Number Title Priority Date Filing Date
PCT/JP2017/035591 Continuation WO2019064526A1 (ja) 2017-09-29 2017-09-29 推定プログラム、推定装置および推定方法

Publications (2)

Publication Number Publication Date
US20200226923A1 US20200226923A1 (en) 2020-07-16
US11462102B2 true US11462102B2 (en) 2022-10-04

Family

ID=65902346

Family Applications (1)

Application Number Title Priority Date Filing Date
US16/827,786 Active 2038-01-06 US11462102B2 (en) 2017-09-29 2020-03-24 Storage medium, estimation device, and estimation method

Country Status (4)

Country Link
US (1) US11462102B2 (ja)
EP (1) EP3690843A4 (ja)
JP (1) JP6939889B2 (ja)
WO (1) WO2019064526A1 (ja)

Citations (13)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPS49132484A (ja) 1973-04-25 1974-12-19
JPH03104774A (ja) 1989-09-20 1991-05-01 Hitachi Ltd 列車混雑度表示方式
JPH10217968A (ja) 1997-02-10 1998-08-18 Toshiba Corp 列車混雑度表示方法及びその表示システム
JP2006188150A (ja) 2005-01-06 2006-07-20 Nippon Signal Co Ltd:The 乗車率予測システム
US20090059793A1 (en) * 2007-08-14 2009-03-05 Greenberg Albert G Traffic engineering method, system and computer program product for managing traffic over dynamic networks during both normal and unexpected traffic scenarios
WO2014203391A1 (ja) 2013-06-21 2014-12-24 株式会社日立製作所 交通行動推定システム、交通システム
WO2015167169A1 (ko) 2014-05-02 2015-11-05 서울대학교 산학협력단 전철 네트워크의 승객 이동경로 추정 방법
CN105046350A (zh) 2015-06-30 2015-11-11 东南大学 基于afc数据的公交客流od实时估计方法
CN105404934A (zh) 2015-11-11 2016-03-16 北京航空航天大学 一种基于条件随机场的城市人口移动数据模型分析方法
JP2016038756A (ja) 2014-08-08 2016-03-22 三菱電機株式会社 エネルギー供給システムの管理・制御支援システム
JP2016168876A (ja) 2015-03-11 2016-09-23 株式会社東芝 混雑予測装置及び混雑予測方法
JP2017010225A (ja) 2015-06-19 2017-01-12 株式会社福山コンサルタント Ｏｄ交通量の実数推定方法、ｏｄ交通量の実数推定装置、ｏｄ交通量の実数推定プログラム、並びに情報記録媒体
JP2017091008A (ja) 2015-11-04 2017-05-25 株式会社日立製作所 データ処理方法およびデータ処理システム

Patent Citations (13)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPS49132484A (ja) 1973-04-25 1974-12-19
JPH03104774A (ja) 1989-09-20 1991-05-01 Hitachi Ltd 列車混雑度表示方式
JPH10217968A (ja) 1997-02-10 1998-08-18 Toshiba Corp 列車混雑度表示方法及びその表示システム
JP2006188150A (ja) 2005-01-06 2006-07-20 Nippon Signal Co Ltd:The 乗車率予測システム
US20090059793A1 (en) * 2007-08-14 2009-03-05 Greenberg Albert G Traffic engineering method, system and computer program product for managing traffic over dynamic networks during both normal and unexpected traffic scenarios
WO2014203391A1 (ja) 2013-06-21 2014-12-24 株式会社日立製作所 交通行動推定システム、交通システム
WO2015167169A1 (ko) 2014-05-02 2015-11-05 서울대학교 산학협력단 전철 네트워크의 승객 이동경로 추정 방법
JP2016038756A (ja) 2014-08-08 2016-03-22 三菱電機株式会社 エネルギー供給システムの管理・制御支援システム
JP2016168876A (ja) 2015-03-11 2016-09-23 株式会社東芝 混雑予測装置及び混雑予測方法
JP2017010225A (ja) 2015-06-19 2017-01-12 株式会社福山コンサルタント Ｏｄ交通量の実数推定方法、ｏｄ交通量の実数推定装置、ｏｄ交通量の実数推定プログラム、並びに情報記録媒体
CN105046350A (zh) 2015-06-30 2015-11-11 东南大学 基于afc数据的公交客流od实时估计方法
JP2017091008A (ja) 2015-11-04 2017-05-25 株式会社日立製作所 データ処理方法およびデータ処理システム
CN105404934A (zh) 2015-11-11 2016-03-16 北京航空航天大学 一种基于条件随机场的城市人口移动数据模型分析方法

Non-Patent Citations (6)

* Cited by examiner, † Cited by third party
Title
EESR—The Extended European Search Report of European Patent Application No. 17926612.7 dated Jul. 10, 2020.
EPOA—Office Action of European Patent Application No. 17926612.7 dated Jun. 1, 2021. References D1-D3 cited in the EPOA were previously submitted in the IDS filed on Oct. 5, 2020.
EPOA—Summons to attend oral proceedings pursuant to Rule 115(1) EPC of European Patent Application No. 17926612.7 mailed on Apr. 4, 2022. References cited in the EPOA were previously submitted in the IDS filed on Oct. 5, 2020.
International Search Report and Written Opinion of the International Searching Authority (Form PCT/ISA/210 & 237), mailed in connection with PCT/JP2017/035591 and dated Jan. 9, 2018 (Total 8 pages).
JPOA—Office Action of Japanese Patent Application No. 2019-544129 dated Dec. 15, 2020 with Full Machine Translation. References cited in the JPOA were previously submitted in the IDS filed on Mar. 24, 2020.
JPOA—Office Action of Japanese Patent Application No. 2019-544129 dated Mar. 16, 2021 with Full Machine Translation. References cited in the JPOA were previously submitted in the IDS filed on Mar. 24, 2020.

Also Published As

Publication number Publication date
JPWO2019064526A1 (ja) 2020-04-02
EP3690843A4 (en) 2020-08-12
JP6939889B2 (ja) 2021-09-22
US20200226923A1 (en) 2020-07-16
WO2019064526A1 (ja) 2019-04-04
EP3690843A1 (en) 2020-08-05

Similar Documents

Publication Publication Date Title
Yang et al. Ensemble learning for short‐term traffic prediction based on gradient boosting machine
US20210005090A1 (en) Method and apparatus for predicting flight delay, device and storage medium
US9696168B2 (en) Travel time information providing apparatus and travel time information providing method
Shabayek et al. A simulation model for the Kwai Chung container terminals in Hong Kong
CN105593769A (zh) 环境感知命令和控制系统
US20160125248A1 (en) Method and service server for providing passenger density information
CN104599002B (zh) 预测订单价值的方法及设备
Bogdan et al. A fractional calculus approach to modeling fractal dynamic games
US11462102B2 (en) Storage medium, estimation device, and estimation method
CN111639591A (zh) 轨迹预测模型生成方法、装置、可读存储介质及电子设备
CN114559989A (zh) 一种列车定位方法、装置、终端设备以及存储介质
Jaipuria et al. A hybrid forecasting technique to deal with heteroskedastic demand in a supply chain
JP2020177354A (ja) 予測装置、予測システム、コンピュータプログラム、予測方法及び生成方法
US11702120B2 (en) Information processing apparatus, information processing method, and non-transitory computer readable medium
Rezazada et al. What influences passenger’s arrival rate at stops in Melbourne: Wasea-Lstm, a novel deep learning model leveraging multi-source data fusion
JP2010061413A (ja) 遅延算出装置及び遅延時間算出方法
Hadas et al. Public transit simulation model for optimal synchronized transfers
Yu et al. Robust online detection in serially correlated directed network
US20220270003A1 (en) Prediction device, prediction method, and prediction program
JP7090816B2 (ja) 問題分解装置、問題分解方法、及び、問題分解プログラム
US20240149444A1 (en) Information processing apparatus, information processing method, and program
CN113077281B (zh) 一种地铁客流量分布预测方法以及设备
CN113256004B (zh) 车辆排班方法、装置、计算机设备及存储介质
Reyes et al. On‐line passenger estimation in a metro system using particle filter
US20220245486A1 (en) Prediction device, prediction method, and prediction program

Legal Events

Date Code Title Description
AS Assignment

Owner name: FUJITSU LIMITED, JAPAN

Free format text: ASSIGNMENT OF ASSIGNORS INTEREST;ASSIGNORS:ENDOH, TOSHIO;MATSUMOTO, KAZUHIRO;SIGNING DATES FROM 20200309 TO 20200311;REEL/FRAME:052209/0589

FEPP Fee payment procedure

Free format text: ENTITY STATUS SET TO UNDISCOUNTED (ORIGINAL EVENT CODE: BIG.); ENTITY STATUS OF PATENT OWNER: LARGE ENTITY

STPP Information on status: patent application and granting procedure in general

Free format text: DOCKETED NEW CASE - READY FOR EXAMINATION

STPP Information on status: patent application and granting procedure in general

Free format text: NON FINAL ACTION MAILED

STPP Information on status: patent application and granting procedure in general

Free format text: RESPONSE TO NON-FINAL OFFICE ACTION ENTERED AND FORWARDED TO EXAMINER

STPP Information on status: patent application and granting procedure in general

Free format text: FINAL REJECTION MAILED

STPP Information on status: patent application and granting procedure in general

Free format text: DOCKETED NEW CASE - READY FOR EXAMINATION

STPP Information on status: patent application and granting procedure in general

Free format text: NOTICE OF ALLOWANCE MAILED -- APPLICATION RECEIVED IN OFFICE OF PUBLICATIONS

STPP Information on status: patent application and granting procedure in general

Free format text: PUBLICATIONS -- ISSUE FEE PAYMENT VERIFIED

STCF Information on status: patent grant

Free format text: PATENTED CASE